Uncovering Additional Predictors of Urothelial Carcinoma from Voided Urothelial Cell Clusters Through a Deep Learning Based Image Preprocessing Technique
Urine cytology is commonly used as a screening test for high grade urothelial carcinoma for patients with risk factors or hematuria and is an essential step in longitudinal monitoring of patients with a prior bladder cancer history. However, the semi-subjective nature of current reporting systems for urine cytology (e.g., The Paris System) can hamper reproducibility. For instance, the incorporation of urothelial cell clusters into the classification schema is still an item of debate and perplexity amongst expert cytopathologists, as several previous works have disputed their diagnostic relevance. Recently, several machine learning and morphometric algorithms have been proposed to provide quantitative descriptors of urine cytology specimens in an effort to reduce subjectivity and include automated assessments of cell clusters. However, it remains unclear how these computer algorithms interpret/analyze cell clusters. In this work, we have developed an automated preprocessing tool for urothelial cell cluster assessment that divides urothelial cell clusters into meaningful components for downstream assessment (i.e., population-based studies, workflow automation). Results indicate that cell cluster atypia (i.e., defined by whether the cell cluster harbored multiple atypical cells, thresholded by a minimum number of cells), cell border overlap and smoothness, and total number of clusters are important markers of specimen atypia when considering assessment of urothelial cell clusters. Markers established through techniques to separate cell clusters may have wider applicability for the design and implementation of machine learning approaches for urine cytology assessment.